Indian Geotech J (OctoberDecember 2015) 45(4):441448
DOI 10.1007/s40098-015-0158-2
ORIGINAL PAPER
Damage Study of a Lightly Stabilised Granular Material Using
Flexural Testing
Dalim Kumar Paul1 Mathanraj Theivakularatnam2 C. T. Gnanendran2
Received: 20 May 2015 / Accepted: 9 July 2015 / Published online: 6 August 2015
Indian Geotechnical Society 2015
Abstract This paper investigates the damage character-
istics of a lightly stabilized granular material using cyclic
load flexural testing with an improved deflection mea-
surement setup. Beam specimens were prepared from a
typical granular base material lightly stabilised with 1.5 %
cement-flyash, cured for 28 days and tested at different
stress ratios (SRs) until the fracture occurred. Fatigue life
was established as the number of cycles required to break
the specimen under stress-controlled cyclic loading at a
frequency of 3 Hz. Fatigue induced damage was evaluated
using the dissipated energy approach. Experimental results
indicated that damage accumulation due to fatigue
approximately follows the Miners linear cumulative
damage rule. The study also found that the accumulated
permanent deformation of lightly stabilised materials
showed good correlation with number of load cycles and
the developed empirical equations at different SRs can be
used to predict the fatigue life. Progressive damage accu-
mulation due to permanent deformation with increases in
the load cycles is also presented in this study.
& Dalim Kumar Paul
dalim49@yahoo.com
Mathanraj Theivakularatnam
m.theivakularatnam@adfa.edu.au
C. T. Gnanendran
r.gnanendran@adfa.edu.au
1 product. In cementitious stabilisation, the use of industrial
Department of Civil and Architectural Engineering, Qatar
University, Doha, Qatar
2 locally available substandard granular materials gives
School of Engineering and Information Technology,
University of New South Wales at ADFA, Canberra,
Australia
Keywords Flexural testing  Lightly stabilised
granular materials  Fatigue  Rutting  Damage
Introduction
The increasing use of heavy vehicles with multiple axles
and wheels and increased tyre pressures to meet the ever
growing demand and need of society has put pressure on
pavement engineers to build roads which can withstand
huge axle tyre pressures and large number of axle load
repetitions. Improving mechanical properties of pavement
materials and to build roads which can handle high axle
tyre pressure through stabilisation has been in practice for
the past few decades. Moreover, problems due to limited
availability of good quality aggregates and the cost asso-
ciated with transporting the materials over long distances to
the construction site has resulted in the solution of using
locally available substandard quality granular materials
with stabilisation. The purpose of stabilisation can range
from simply providing a working platform to significantly
improving the mechanical properties of the road base
material. Cementitious stabilization has also been used as
an effective and economic option for the rehabilitation of
old existing pavements and in road widening/upgrading
projects.
Cementitious stabilisation refers to the addition of
cementitious binders such as cement, flyash, lime and slag
which forms cementitious product as a result of chemical
reaction and binds the aggregates together to form a better
by-products such as flyash and slag as well as the use of
benefit to the environment apart from improved strength,
stiffness and durability of the stabilised granular material.
123
442
Depending on the type, amount and enhancing quality of
a binder, the chemically stabilised material can be either
fully bounded material or partially bounded material [7,
23]. Fully bounded material which is also referred to as
cemented material generally made by the addition of larger
percentage of cement binder (e.g. [5 % cement) to the
granular material has significant tensile strength and it is
characterised based on its flexural properties. However, the
cost and possible shrinkage cracks associated with a
cemented material have limited the amount of stabilizers in
places where these two limitations are real concerns [12].
Modification of granular materials by adding small amount
(e.g. \5 %) of cementitious binders, known as light sta-
bilisation, could be a low cost rehabilitation method that is
practically useful for constructing road pavements over
expansive and/or weak subgrade and sub-base materials.
Moreover, the lower amount of binder also eliminates the
risk of shrinkage cracking. A lightly stabilized pavement
base offers stiffer, more uniform and more water resistant
base layer than an unstabilized base. Its increased strength
and stiffness provides an excellent support for low or
medium volume roads constructed with thin sprayed seal or
with a thin asphalt layer.
The Austroads [7] characterises lightly stabilised gran-
ular material in the same way as an unbound granular
primarily based on resilient modulus obtained using repe-
ated load triaxial test. Light stabilisation produces a brittle
mix and brings in tensile capacity to the unbound granular
material; however, the resultant material is weak in tension
due to low percentage of binder addition. When used as a
road base material subjected to moving traffic loads, bot-
tom of the lightly stabilised semi-rigid material is imper-
illed to repeated tensile stress/strain which causes damage.
As the number of load application increases, depending on
the magnitude of the loadings and the environmental con-
ditions, cumulative effect of damage may eventually result
in the development and propagation of tensile cracks
leading to fatigue fracture.
Limited information is available in the literature on the
damage characteristics of lightly stabilised materials.
Gnanendran and Piratheepan [14] used cyclic load indirect
diametrical tensile (IDT) testing to evaluate fatigue beha-
viour of granular base materials stabilised with 35 % slag-
lime binder and developed fatigue relationships in terms of
stress and strain ratios. However, Paul and Gnanendran
[24] showed that the tensile characterization of lightly
stabilized materials can also be carried out using flexural
beam testing. Compared with IDT, flexural beam testing is
preferred for fatigue damage assessment because of the
similarities in the stress conditions in the beam sample
under the loading arrangements used in that test and the
base layer of a pavement structure under in situ traffic
wheel loading. Therefore one of the objectives of this paper
123
Indian Geotech J (OctoberDecember 2015) 45(4):441448
is to study the progressive accumulation of damage under
cyclic loading using flexural beam testing.
Moreover, rutting failure is a likely distress mode for
lightly stabilised material similar to unbound granular
materials and subgrade materials. Permanent deformation,
although small in magnitude, also accumulates in the
material in each load cycle and ultimately contributing to
the total permanent deformation (or rutting) of the flexible
pavement structure. A typical laboratory permanent
deformation test procedure consists of applying repeated
constant level of stress to each specimen in a cyclic load
triaxial test. Usually, the number of repeated loading cycles
goes beyond 100,000 cycles which enables the study of
long term rutting behavior. However, in this study, per-
manent deformation measured in the cyclic load flexural
beam test with different stress ratios (SRs) were analyzed
against number of load cycles. The fatigue and rutting
studies emphasize the fact that the materials of a pavement
structure should be characterized and designed in such a
way that they could withstand against fatigue and rutting
failure during its span of design life.
Experimental Investigation
Experimental program described in the following sections
includes selection of materials, determination of mixture
characteristics, sample preparation and testing in flexure.
Materials
The parent base material used in this investigation was the
freshly quarried granular material obtained from the
Mugga Quarry in Canberra, and supplied by Boral
Resources Pty. It was geologically identified as rhyodacite
porphyry (an acid sub-volcanic igneous rock). According
to the Unified Soil Classification System, the material was
classified as well-graded sandy gravel with some fines, and
satisfied the grading ranges guidelines for type 1 gradation
C road base materials according to ASTM D124107 [4].
The grain size distribution of this material is shown in
Fig. 1.
Obtaining consistent samples from such a crushed rock
material is very difficult andnon-uniform proportions of
particle in the samples are common [27]. To overcome this
inconsistency, reconstituted material with an unchanged (or
consistent) material grading (shown in Fig. 1) were adop-
ted for both the materials. This reconstituted materials, of
uniform grading all over the test regime, has been achieved
by sieving a large batch of parent material, separating them
into different particle sizes and then remixing them at
suitable weight proportions. Therefore the adopted grading
Indian Geotech J (OctoberDecember 2015) 45(4):441448
Fig. 1 Particle size distribution curve of parent material
Table 1 Physical properties of parent material
Physical property
Wet strength (kN)
Dry strength (kN)
Liquid limit (%)
Plastic limit (%)
Plasticity index (%)
Linear shrinkage (%)
Optimum moisture content (%)
Maximum dry density (kg/m3)
for the reconstituted sample, hereafter referred as the par-
ent material, was essentially the same for all the samples
that were tested in this experimental investigation. Physical
properties of the parent material are presented in Table 1.
The stabilizers chosen for this experimental investiga-
tion were general blend (GB) cement and fly ash. The GB
cement and fly ash were used in the ratio of 7525 % by
dry weight. Previous investigations [25, 26] indicated that
parent material with a binder content equal to 1.53 %
cement-fly ash typically performs as lightly stabilised
materials. Therefore, an arbitrary binder content of 1.5 %
cement-fly ash was chosen to ensure the mixture is lightly
stabilised materials.
Specimen Preparation
Dry density and moulding moisture content are two
important factors which significantly affect the perfor-
mances of pavement materials. Modified proctor
443
Fig. 2 Dry density-moisture content relationship of lightly stabilised
materials
compaction is practically used to determine MDD and
OMC. However, sometimes it is quite difficult to get the
Value soil compacted as tightly as the modified Proctor test, and
also it takes longer time to do it. Therefore, the standard
157 Proctor test is often used as an alternative. In this study,
217 Standard Proctor test was chosen to compact the mixture
18 and the test was carried out to establish the dry density-
15 moisture content relationship according to AS-1289.5.1.1.
3 [2] which is similar to ASTM D698 [3]. Figure 2 shows the
n/a dry-density-moisture content curve of the materials/mixes.
9.0 The maximum dry density (MDD) of the parent material
2092 without any binder was found to be 2088 kg/m3 whereas
the addition of 1.5 % cement-flyash increased this value to
2099 kg/m3 with its optimum moisture content (OMC)
remaining almost constant at 9 %. Therefore, all the sam-
ples stabilized with 1.5 % binder content were prepared
with a fixed moisture content of 9 % throughout this study.
Flexural beam specimens of dimension 285 mm 9
76 mm 9 76 mm were prepared according to ASTM D
1632-07 [6]. Initially, an oven-dried granular material was
proportioned based on the reconstituted particle size dis-
tribution and then dry-mixed with binder before being
thoroughly mixed with water for about 5 min. The beam
mold was set up with its top platen open and molding oil
was applied for the easy removal. Beams were prepared by
pouring the mix in detachable steel moulds in three equal
layers each of them was manually compacted with 90
blows using a square-ended tamping rod at a target unit
weight (99.5 % of the MDD). Each layer was scarified for
proper bonding with the next layer. After levelling the mix,
the top plate was placed on top of it and fitted with it. Then
the beam was compacted by the application of a static
compression load using the universal testing machine until
123
444 Indian Geotech J (OctoberDecember 2015) 45(4):441448

curing period, they were tested in flexure immediately after

being taken from the fog room.

Flexural Fatigue Testing and Data Acquisition

Beams were tested in bending under third-point loading

Fig. 3 Compaction of flexural beam by static compression load
the desired height of 76 mm was achieved. Figure 3 shows
the static compaction process used in this investigation.
The prepared samples were allowed to cure in the mould
for 24 h until they gained sufficient strength for the mould
to be removed without any damage to the sample. After
removal from the mould, the samples were wrapped with
polythene and placed in a fog room at 23 2 C and
95 5 % humidity for curing. At the end of 28 days
Fig. 4 Photographic view of the experimental setup used in flexural
testing
configuration over a span length of 228 mm. In conven-
tional flexural testing [13, 16, 20, 10, 1], mid-span vertical
deformation is generally measured externally which may
include some extraneous deformation. This extraneous
movement may give rise to an overestimation of the
flexural strain thus underestimating of the stiffness mod-
ulus/fatigue life, which could lead to uneconomical and
overly conservative thickness design of pavement. To
eliminate these erroneous vertical movements, several
researchers [17, 18] adopted special measurement tech-
nique to measure the vertical deflection of composite
beams. In this investigation, a new flexural testing setup
similar to the one proposed in ASTM C 1609-10 [5] but
with further improved instrumentation was used to mon-
itor the net mid-span vertical deflection of beam
specimens.
Figures 4 and 5 show the flexural test setup used in this
study. Mid-span vertical deflection was measured by two
LVDTs held by two horizontal bars (cross-sectional areas
of 20 mm 9 15 mm)one on each side of the beam
specimen. A pair of C-shaped aluminium brackets was
used to support the horizontal bars. To prevent bending of
the bar, it was pin jointed at one end and roller supported at
the other (i.e., at one end, the bar was held by bolts around
which they could rotate but not move horizontally or ver-
tically with respect to a point on the neutral axis of the
sample and at the other end it rested on another bolt so that
the bar could slide horizontally but not move vertically).
The LVDT tip was supported on a thin L-shaped angle
glued to the top surface of the beam specimen. The average
of the two LVDT readings was taken as the mid-span
vertical deflection of the test specimen. Therefore it was

Fig. 5 Schematic diagram of P Steel rod

the setup for flexural testing
Spherically seated block
Beam specimen
Steel ball 76mm*76mm*285
P/2 P/2
L-shaped angle C-shaped
bracket
C-shaped
bracket Pin joint for the
horizontal line bar
LVDT Horizontal line bar
Steel rod Steel ball

76 mm 76 mm 76 mm

123
Indian Geotech J (OctoberDecember 2015) 45(4):441448 445

possible to measure accurately and reliably the centre

deflection of the beam from the neutral axis. The schematic
of the testing setup is shown in Fig. 5. The use of on-
sample deflection measurement setup detailed above
enabled to overcome any unexpected beam crushing on the
supports as well as movement/tilting of the LVDT as a
result of being attached as a propped cantilever to the
middle horizontal bar as could be seen in ASTM C 1609-10
[5].
Apart from the newly developed experimental setup, it
is also important that the load and the deformation readings
are measured and acquired to the acceptable level of
accuracy. The loading was applied through Moogs testing
machine which comprised of a hardware called Moogs
servo controller and a software known as moog axis control
software (MACS). All the input parameters defining load-
ing pattern (minimum, maximum loadings) and frequency Fig. 6 Typical stressstrain response at failure (SR = 0.85)
were stored in MACS which then controlled the systems
hardware. A load cell with the capacity of 22.24 kN
Table 2 Flexural fatigue test results
(5000 lbs) was selected for the experiment which was
periodically calibrated by Australian calibration services. Binder content SR MOR Cyclic applied stress Fatigue
(%) (%) (MPa) (MPa) life, Nf
Two miniature LVDTs with -3 mm to ?3 mm displace-
ment range were used to measure the flexural deformation 1.5 50 0.25 0.125 327,797
which were calibrated prior to test. A data acquisition 1.5 60 0.25 0.150 195,565
program LabVIEW was used to collect data from both 1.5 70 0.25 0.175 100,021
LVDTs and load cell at a rate of 500 Hz and the data 1.5 80 0.25 0.200 18,000
showed that selected rate is high enough to capture the 1.5 85 0.25 0.213 41
peak loads and deformations.
Initially monotonic flexural testing was performed to
determine the modulus of rupture (MOR). Paul and Gna-
nendran [24] found that the loading rate has a profound cycle equal to zero could lead to rocking action and the
effect on the strength and stiffness of a lightly stabilized range of the stress/load would have an effect on the
material and recommended 0.5 mm/min as a suitable dynamic properties of the material [21]. Figure 6 shows a
loading rate. Therefore, the loading was applied using a typical stressstrain response of the lightly stabilised
Moog testing machine at the rate of 0.5 mm/min. The material at failure.
MOR defined by Eq. 1 is the maximum tensile stress
produced at the outermost bottom fiber of a specimen
subjected to flexural loading. Analysis and Discussion of Test Results
PL
rf 1 Resistance to fatigue is one of the most important consid-
bd 2 erations in the design of a pavement involving lightly
where rf = flexural stress in MPa, P = peak load in N, stabilised materials. There are different definitions of fati-
b = average width of specimen in mm, L = span length in gue failure based on a materials responses as well as
mm and d = average depth of specimen in mm. energy approaches. Complete fracture of a specimen is
Fatigue testing was then conducted on another batch of often used as an indication of fatigue failure in stress-
identical specimens at a particular stress ratio (SR = ap- controlled flexural fatigue testing [9, 28]. The current study
plied stress/MOR) until the failure of the specimen. This adopts the breakage of specimens as fatigue failure criteria
was a stress-controlled sinusoidal-type loading test with a and hence the number of cycles required to fail the speci-
frequency of 3 Hz. All the specimens were conditioned by men is defined as the fatigue life (Nf). Table 2 and Fig. 7
applying 1000 cycles at a stress ratio of 40 % prior to show the fatigue life of lightly stabilised materials for
fatigue testing. It should be noted that a small amount of different SRs. As can be observed, the fatigue life is
constant load (e.g. corresponding to 10 % of the MOR) was decreasing with the increases in SRs (i.e., applied cyclic
kept as a seating load since the minimum cyclic load for a stress).

123
446
Fig. 7 Variations in fatigue life with stress ratio
One popular method for modelling fatigue life is the so-
called SN curve (i.e., the SR versus Nf curve). The fatigue
lives were plotted against SRs to establish the SN curve
and a linear relationship between the SR and Nf was
obtained (Fig. 6), unlike the traditionally nonlinear varia-
tion obtained by many researchers [22, 29]. The SN
relationship obtained for the specimens investigated in this
study may be expressed by Eq. 2.
Nf 7:75  105 SR  9:38  105 R2 0:97 2
The major distress modes of a stabilised pavement layer
caused by cyclic loading are fatigue and permanent
deformation. Damage accumulation caused by fatigue
and permanent deformation was estimated in this study.
Fatigue induced damage is often evaluated in terms of
damage index (Di). A damage index (Di) can be defined as
[15]
PkN
W
Di k1 3 permanent deformation components. The permanent ver-
WT
P
where W = accumulated dissipated energy at cycle
N and WT = the total energy dissipation capacity (WT) of
the material at a given stress level.
The damage indices for the lightly stabilised granular
mix for varying SR values were plotted against the cycles
ratio (i.e., applied cycles divided by the number of cycles
to failure, N/Nf), as shown in Fig. 8. This figure also
illustrates the Miners linear damage accumulation rule
[19] which has the following expression for variable stress
levels
X Ni
1 4
Nf ;i
where Ni is the number of cycles at stress level i and
Nf,i = total number of cycles to failure at stress level
i. According to Miners hypothesis, a material will fail
when the summation of the damage ratio shown on the left-
123
Indian Geotech J (OctoberDecember 2015) 45(4):441448
Fig. 8 Progressive damage accumulation with load cycles
hand side of Eq. 4 equals 1. It can be observed from Fig. 8
that the lightly stabilised granular mix approximately
conforms to the Miners rule at various stress levels.
Understanding the accumulated damage sustained by a
pavement material layer up to a certain stage of its life (as
obtained from Fig. 8) is important because its remaining
life can be determined from this information and a suitable
rehabilitation strategy can be undertaken.
Vertical permanent deformation directly contributes to
the total rutting of a pavement structure and is regarded as
one of the causes of stress-induced damage. To study the
rutting potential of the lightly stabilised material, the
accumulation of permanent deformation was also deter-
mined by monitoring the mid-span vertical deflection of the
beam specimens. The average total vertical deformation in
each cycle consisted of both an elastic and a plastic or
tical deformation accumulation (Dp) at any cycle, N, was
normalised with respect to the total accumulated permanent
deformation up to failure (Dp,f) and was plotted against the
cycles ratio (N/Nf).
The permanent deformation ratio (Y = Dp/Dp,f) versus
cycles ratio (X = N/Nf) plots for different SRs exhibited
linear variations in the loglog scale (Fig. 9) from which
the permanent deformation at a given N could be predicted.
The regression equations obtained for the relationships
between Y and X that could be used for estimating the
permanent deformation over a certain number of load
cycles established from this study are also given in Fig. 9.
It is noted that various researchers (e.g. [8, 11]) have
suggested similar loglog variations between permanent
deformation and load cycles to investigate the damage
induced by permanent deformation and this study on
lightly stabilised granular material also supports that
finding.
Indian Geotech J (OctoberDecember 2015) 45(4):441448
Fig. 9 Variations in permanent deformation accumulation with load cycles for different SRs
Conclusions
The use of light stabilisation using small percentage of
cementitious binders such as cement, flyash, lime and slag
is gaining popularity as an effective and economic option
in the construction of a new or rehabilitation of an existing
pavement. However, long-term performance of lightly
stabilised base is still a concern to pavement engineers,
447
especially the damage associated due to fatigue and rutting
under repeated traffic loading. This paper investigated the
progressive damage due to fatigue and accumulation of
permanent deformation of lightly stabilised materials using
flexural testing under cyclic loading.
Beam specimens were prepared from a typical granular
material lightly stabilised with 1.5 % cement-flyash binder
content and tested under monotonic flexural testing with
123
448
improved instrumentation to determine MOR. Then, cyclic
load testing was performed on another set of specimens at
various stress levels until the fatigue fracture occurred.
Fatigue life was estimated as the number of load cycles
required to break the specimen. Test results indicated a
linear relationship of the fatigue life with SR and empirical
correlation between them for the prediction of the fatigue
life of a lightly stabilised material was proposed. It was
also found that damage accumulation due to fatigue
approximately follows the Miners rule. Accumulation of
permanent deformation in a lightly stabilised material
under traffic type cyclic loading was also investigated and
empirical damage models applicable for different SRs were
developed to predict the permanent deformation at various
stages of its design life.
Acknowledgments The authors would like to thank Mr. David
Sharp, Mr. Jim Baxter and Mr. Mathew Barret for their technical
assistance during the experimental work reported in this paper.
References
1. Arulrajah A, Disfani MM, Haghighi H, Mohammadinia A, Hor-
pibulsuk S (2015). Modulus of rupture evaluation of cement
stabilized recycled glass/recycled concrete aggregate blends.
Constr Build Mat 84:146155
2. AS 1289.5.1.1 (2003) Methods of testing soils for engineering
purposes, method 5.1.1: soil compaction and density testsde-
termination of the dry density/moisture content relation of a soil
using standard compactive effort. Standards Australia, Sydney
3. ASTM D69807 (2007) Test method for laboratory compaction
characteristics of soil using standard effort [12 400 ft-lbf/ft3 (600
kN-m/m3)], annual book of ASTM Standards, vol 04.08. ASTM
International, Conshohocken
4. ASTM D124107 (2007) Specification for materials for soil-ag-
gregate subbase, base and surface courses, annual book of ASTM
Standards, vol 04.08. ASTM International, Conshohocken
5. ASTM C160910 (2010) Standard test method for flexural per-
formance of fiber-reinforced concrete (using beam withn third-
point loading), annual book of ASTM Standards, vol 04.08.
ASTM International, Conshohocken
6. ASTM D163207 (2007) Standard practice for making and cur-
ing soilcement compression and flexure test specimens in the
laboratory, annual book of ASTM Standards, vol 04.08. ASTM
International, Conshohocken
7. Austroads (2010) Guide to pavement technologyPart 2: pave-
ment structural design. AUSTROADS, Sydney
8. Behzadi G, Yandell WO (1996) Determination of Elastic and
Plastic Subgrade Soil Parameters for Asphalt Cracking and Rut-
ting Prediction, vol 1540. Transportation Research Board,
Washington, pp 97104
9. Bhogal BS, Coupe PS, Davies J, Fendukly LM (1995) Dynamic
flexure tests of soil-cement beams. J Mater Sci Lett 14:302304
10. Disfani MM, Arulrajah A, Haghighi H, Mohammadinia A, Hor-
pibulsuk S (2014) Flexural beam fatigue strength evaluation of
crushed brick as a supplementary material in cement stabilized
recycled concrete aggregates. Constr Build Mat 68:667676
11. Diyaljee VA, Raymond GP (1982) Repetitive load deformation
of cohesionless soil. J Geotech Eng Div ASCE 108(10):
12151229
123
Indian Geotech J (OctoberDecember 2015) 45(4):441448
12. Foley G, Australian Stabilisation Expert Group (2001) Contract
reporteffect of design, construction and environmental factors
for long-term performance of stabilised materials, Report No.
RC91022-1, Austroads, Sydney, Australia
13. Fu P, Jones D, Harvey JT, Bukhari SA (2009) Laboratory test
methods for foamed asphalt mix resilient modulus. Road Mater
Pavement Des 10(1):187212
14. Gnanendran CT, Piratheepan J (2010) Determination of fatigue
life of a granular base material lightly stabilized with slag lime
from indirect diametral tensile testing. J Transp Eng 136(8):
736745
15. Grzybowski M, Meyer C (1993) Damage accumulation in con-
crete with and without fiber reinforcement. ACI Mater J 90(6):
594604
16. Khoury NN, Zaman MM (2006) Durability effects on flexural
behaviour of fly ash stabilised limestone aggregate. J Test Eval
34(3):167175
17. Kim DJ, Naaman AE, El-Tawil S (2008) Comparative flexural
behaviour of four fiber reinforced cementitious composites.
Cement Concr Compos 30(10):917928
18. Lin C, Kayali O, Morozov EV, Sharp DJ (2011) Deflection
hardening of steel fibre reinforced cementitious composites with
high volume fly ash. In: Proceedings of the 9th International
Symposium on HPC, Rotorua, New Zealand, pp 18
19. Miner MA (1945) Cumulative damage in fatigue. J Appl Mech T
ASME 12(3):A159A164
20. Moffat MA, Yeo EEY (1998) Relationship between unconfined
compressive strength and flexural modulus for cemented mate-
rials. TR Working Document No. R-98/024, ARRB, Melbourne,
Australia
21. Murdock JW, Kesler CE (1958) Effect of range of stress on
fatigue strength of plain concrete beams. ACI J 55(2):221232
22. NCHRP (2004) Final reportguide for mechanistic-empirical
design of new and rehabilitated pavement design structurespart
2 design inputschapter 2: material characterization, ERES
Division of ARA, Inc., Champaign
23. Paul D, Gnanendran C (2012) Prediction of nonlinear stress
strain relationship of lightly stabilized granular materials from
unconfined compression testing. J Mater Civ Eng 24(8):
11181124
24. Paul DK, Gnanendran CT (2012b) Characterisation of lightly
stabilised granular base materials by flexural beam testing and
effects of loading rate. ASTM Geotech Test J 35(5)
25. Paul DK, Gnanendran CT (2013) Stressstrain behaviour and
stiffness of lightly stabilised granular materials from UCS testing
and their predictability. Int J Pavement Eng 14(3):291308
26. Paul DK, Gnanendran CT, Piratheepan J (2010) Determination of
stiffness properties of lightly stabilised granular materials from
IDT testing using numerical analysis. In: Proceeding of the 17th
Southeast Asian Geotechnical Conference, Taiwan Geotechnical
Society and Southeast Asian Geotechnical Society, vol 1,
pp 7578
27. Paul DK, Gnanendran CT (2011) Effect of loading rate on the
properties of lightly stabilized granular materials by using flex-
ural beam testing. In: 14th Pan-AmericanCGS Conference on
Soil Mechanics and Geotechnical Engineering, Pan-Am CGS
2011 Organizing Committee, 26 October, 2011, Toronto,
Canada
28. Sobhan K, Das BM (2007) Durability of soil-cements against
fatigue fracture. J Mater Civ Eng 19(1):2632
29. Swanson TE, Thompson MR (1967) Flexural fatigue strength of
lime-soil mixtures, vol 198. Highway Research Records, High-
way Research Board, National Research Council, Washington,
pp 9-18